Scholarlink

Secure Neural Network Inference (SNNI) protocols, vital for privacy-preserving AI, face substantial computational and communication overhead. Dynamic Early-Exit (EE) networks could help decrease the overhead, but existing SNNI protocols do not support such networks. We introduce QUOKKA, the first system to enable SNNI for confidence-based EE neural networks using secure Multi-Party Computation. QUOKKA addresses the challenges of dynamic decision-making and sensitive intermediary result handling in SNNI. Implemented with EENet and CrypTen, QUOKKA achieves 2–5 \(\times \) acceleration over traditional SNNI without a decrease in accuracy. Our findings demonstrate practical, highly efficient, and privacy-preserving SNNI for dynamic AI, paving the way for broader Machine-Learning-as-a-Service deployment.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

QUOKKA: Faster Secure Neural Network Inference with Early-Exit Technology

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Daphnee Chabal,
  • Dolly Sapra,
  • Cees de Laat,
  • Zoltán Ádám Mann

摘要

Secure Neural Network Inference (SNNI) protocols, vital for privacy-preserving AI, face substantial computational and communication overhead. Dynamic Early-Exit (EE) networks could help decrease the overhead, but existing SNNI protocols do not support such networks. We introduce QUOKKA, the first system to enable SNNI for confidence-based EE neural networks using secure Multi-Party Computation. QUOKKA addresses the challenges of dynamic decision-making and sensitive intermediary result handling in SNNI. Implemented with EENet and CrypTen, QUOKKA achieves 2–5 \(\times \) acceleration over traditional SNNI without a decrease in accuracy. Our findings demonstrate practical, highly efficient, and privacy-preserving SNNI for dynamic AI, paving the way for broader Machine-Learning-as-a-Service deployment.